© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-RES-MATRYOSHKA-DC
    7. 1637
    Prev
    Next
    INDEX
    Explanations

    phrases used in consumer reviews of products, including value judgements, service experiences, and overall recommendations.

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.22.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    chanind/pile-uncopyrighted-gemma-1024-abbrv-1B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    unie
    -0.60
    esModule
    -0.60
     nonatomic
    -0.51
    vido
    -0.50
    RTDA
    -0.50
     fearless
    -0.49
    zám
    -0.49
    oznam
    -0.48
     getItemCount
    -0.48
     contactez
    -0.47
    POSITIVE LOGITS
     mediocre
    1.79
     poor
    1.58
     mediocrity
    1.50
     disappointing
    1.49
     underwhelming
    1.43
    poor
    1.37
     lousy
    1.36
     unsatisfactory
    1.33
     dismal
    1.32
     crappy
    1.31
    Activations Density 1.533%

    No Known Activations